Tornado Data Analysis in the United States

Chathurika Narayana

2023-12-05

Introduction

Here are the goals of the data explore:

1. Explore the Tornado Dataset

Data Cleaning and Pre-processing

Step I. Identify the data information

Name Description Type Units RangeValues
Tornado Number (om) An identifier for each tornado within a specific year Integer
Year (yr) The year in which the tornado occurred Integer 1950 - 2022
Month (mo) The month of the tornado occurrence Integer 1 to 12
Day of the Month (dy) The specific day of the month when the tornado struck Integer 1 to 31
Date The date of the tornado occurrence categorical MMDYYYY
Time The time of day when the tornado was reported categorical HHMMSS
Timezone (tz) The timezone, using the canonical tz database Integer
State(st) Affected State categorical A two-letter postal abbreviation representing the affected state. (DC=Washington,DC;PR=Puerto Rico,VI=Virgin Islands)
State FIPS Number (stf) The State FIPS (Federal Information Processing Standards) number. Integer
Magnitude (mag) The magnitude of the tornado, recorded on the F scale. EF scale from 2007. Some are estimated Integer
Injuries (inj) The number of injuries associated with the tornado Integer
Fatalities (fat) The number of fatalities caused by the tornado Integer
Loss (loss) Estimated property loss information, with values grouped into ranges for years before 1996 numeric In dollars Prior to 1996, values were grouped into ranges. The reported number for such years is the maximum of its range
Starting Latitude (slat) The geographical coordinates of the tornado’s starting point. (Starting latitude ) numeric Degrees
Starting Longitude (slon) The geographical coordinates of the tornado’s starting point. (Starting longitude) numeric Degrees
Ending Latitude (elat) The geographical coordinates of the tornado’s endpoint. (Ending latitude) numeric Degrees
Ending Longitude (elon) The geographical coordinates of the tornado’s endpoint. (Ending longitude) numeric Degrees
Length (len) The length of the tornado’s path numeric In miles
Width (wid) The width of the tornado’s path Integer In yards
Number of States Affected (ns) The number of states impacted by the tornado Integer 1, 2, or 3..
State Number (sn) The state no of the row Integer Indicates whether the row contains the entire track information for a state (1) or if there is at least one more entry for the state for that tornado (0)
FIPS Codes for Counties (f1, f2, f3, f4) FIPS codes for up to four counties affected by the tornado Integer
Was the Magnitude Estimated (fc) A logical variable indicating whether the magnitude value was estimated Integer 1 or 0

1. Explore the Tornado Dataset..cont

Step II. Check for missing values in the data set

##    om    yr    mo    dy  date  time    tz    st   stf   stn   mag   inj   fat 
##     0     0     0     0     0     0     0     0     0     0     0     0     0 
##  loss closs  slat  slon  elat  elon   len   wid    ns    sn    sg    f1    f2 
##     0     0     0     0     0     0     0     0     0     0     0     0     0 
##    f3    f4    fc 
##     0     0     0

Step III. Identify and Handling Inconsistent Values

##        om               yr             mo               dy       
##  Min.   :     1   Min.   :1950   Min.   : 1.000   Min.   : 1.00  
##  1st Qu.:   285   1st Qu.:1976   1st Qu.: 4.000   1st Qu.: 8.00  
##  Median :   588   Median :1995   Median : 6.000   Median :16.00  
##  Mean   :113189   Mean   :1992   Mean   : 5.968   Mean   :15.93  
##  3rd Qu.:  1117   3rd Qu.:2008   3rd Qu.: 7.000   3rd Qu.:24.00  
##  Max.   :622080   Max.   :2022   Max.   :12.000   Max.   :31.00  
##      date               time                 tz             st           
##  Length:68701       Length:68701       Min.   :0.000   Length:68701      
##  Class :character   Class :character   1st Qu.:3.000   Class :character  
##  Mode  :character   Mode  :character   Median :3.000   Mode  :character  
##                                        Mean   :3.001                     
##                                        3rd Qu.:3.000                     
##                                        Max.   :9.000                     
##       stf             stn              mag               inj          
##  Min.   : 1.00   Min.   :  0.00   Min.   :-9.0000   Min.   :   0.000  
##  1st Qu.:18.00   1st Qu.:  1.00   1st Qu.: 0.0000   1st Qu.:   0.000  
##  Median :28.00   Median : 11.00   Median : 1.0000   Median :   0.000  
##  Mean   :29.22   Mean   : 23.17   Mean   : 0.6711   Mean   :   1.419  
##  3rd Qu.:42.00   3rd Qu.: 31.00   3rd Qu.: 1.0000   3rd Qu.:   0.000  
##  Max.   :78.00   Max.   :232.00   Max.   : 5.0000   Max.   :1740.000  
##       fat                loss               closs               slat      
##  Min.   :  0.0000   Min.   :0.000e+00   Min.   :       0   Min.   :17.72  
##  1st Qu.:  0.0000   1st Qu.:0.000e+00   1st Qu.:       0   1st Qu.:33.18  
##  Median :  0.0000   Median :0.000e+00   Median :       0   Median :37.00  
##  Mean   :  0.0893   Mean   :1.167e+05   Mean   :     914   Mean   :37.13  
##  3rd Qu.:  0.0000   3rd Qu.:4.000e+00   3rd Qu.:       0   3rd Qu.:40.92  
##  Max.   :158.0000   Max.   :1.550e+09   Max.   :12250000   Max.   :61.02  
##       slon              elat            elon              len         
##  Min.   :-163.53   Min.   : 0.00   Min.   :-163.53   Min.   :  0.000  
##  1st Qu.: -98.42   1st Qu.: 0.00   1st Qu.: -94.78   1st Qu.:  0.120  
##  Median : -93.56   Median :32.55   Median : -84.72   Median :  0.800  
##  Mean   : -92.76   Mean   :22.96   Mean   : -56.83   Mean   :  3.489  
##  3rd Qu.: -86.73   3rd Qu.:38.65   3rd Qu.:   0.00   3rd Qu.:  3.200  
##  Max.   : -64.72   Max.   :61.02   Max.   :   0.00   Max.   :234.700  
##       wid               ns              sn               sg          f1       
##  Min.   :   0.0   Min.   :1.000   Min.   :0.0000   Min.   :1   Min.   :  0.0  
##  1st Qu.:  20.0   1st Qu.:1.000   1st Qu.:1.0000   1st Qu.:1   1st Qu.: 37.0  
##  Median :  50.0   Median :1.000   Median :1.0000   Median :1   Median : 85.0  
##  Mean   : 107.8   Mean   :1.009   Mean   :0.9914   Mean   :1   Mean   :104.6  
##  3rd Qu.: 100.0   3rd Qu.:1.000   3rd Qu.:1.0000   3rd Qu.:1   3rd Qu.:137.0  
##  Max.   :4576.0   Max.   :3.000   Max.   :1.0000   Max.   :1   Max.   :810.0  
##        f2                f3                f4                 fc         
##  Min.   :  0.000   Min.   :  0.000   Min.   :  0.0000   Min.   :0.00000  
##  1st Qu.:  0.000   1st Qu.:  0.000   1st Qu.:  0.0000   1st Qu.:0.00000  
##  Median :  0.000   Median :  0.000   Median :  0.0000   Median :0.00000  
##  Mean   :  8.604   Mean   :  1.687   Mean   :  0.5067   Mean   :0.02712  
##  3rd Qu.:  0.000   3rd Qu.:  0.000   3rd Qu.:  0.0000   3rd Qu.:0.00000  
##  Max.   :820.000   Max.   :710.000   Max.   :507.0000   Max.   :1.00000

1. Explore the Tornado Dataset..cont

Issues/considerations

  1. “mag” (Magnitude) Column:

Identify rows which have negative magnitude.

## [1] "No of rows where magnitude is negative: 756"

Identify years where magnitude is negative.

## [1] 2016 2017 2018 2019 2020 2021 2022

1. Explore the Tornado Dataset..cont

  1. “date” and “time” Columns:
## [1] "2023-12-05 21:00:00 CST" "2023-12-05 02:15:00 CST"
## [3] "2023-12-05 02:20:00 CST" "2023-12-05 04:00:00 CST"
## [5] "2023-12-05 07:30:00 CST" "2023-12-05 17:00:00 CST"
  1. “st” Column:
##   st     state_full
## 1 OK       Oklahoma
## 2 NC North Carolina
## 3 KY       Kentucky
## 4 KY       Kentucky
## 5 MS    Mississippi
## 6 PA   Pennsylvania

3. Tornado Impact Assessment

3.1 Assess the human impact of tornadoes

  1. Interactive line plot to visualize the trends in fatalities and injuries over the years.
  • Injuries and Fatalities have seen fluctuations, with some years experiencing higher counts than others implies the variability.

3. Tornado Impact Assessment cont..

  1. an interactive bar plot to visualize the total injuries and fatalities for each EF rating.
  • EF4 tornadoes and EF3 tornadoes contribute significantly to total injuries and fatalities, highlighting the potential severity of these tornado categories.

3. Tornado Impact Assessment cont..

  1. interactive map using leaflet to visualize state-wise tornado Injuries and Fatalities.

3. Tornado Impact Assessment cont..

3.2 Assess the economic impact of tornadoes

  1. A line plots to show the trends in property and crop losses over the years.

  • Property losses show a fluctuating pattern over the years, with some noticeable spikes.
  • Crop losses seem to vary, but there are years with significant increases.
  • The total economic impact follows a similar pattern to property losses.

3. Tornado Impact Assessment cont..

  1. Assess Economic Impact by State

Bar plots which have state-wise highest Property loss and Crop loss.

Top States with Highest Property Loss:

  • Texas, Tennessee, Ohio, Iowa, and Louisiana are among the states with the highest property losses.
  • The bar plot provides a clear comparison of the magnitude of property losses across these states.

Top States with Highest Crop Loss:

  • Mississippi, Georgia, Nebraska, North Dakota, and Minnesota are prominent in terms of crop losses.
  • The plot emphasizes the variability in crop losses among the top states.

3. Tornado Impact Assessment cont..

  1. Interactive map for state-wise total economic loss including Property loss and Crop loss.
  • Darker shades indicate higher total economic losses, offering a quick overview of the most affected states.
  • Hovering over each state provides detailed information on the total loss, property loss, and crop loss.

Shinny Application

Conclusion